Multiprocessing computer system employing a cluster communication error reporting mechanism

• US 6,401,174 B1
• Filed: 09/04/1998
• Issued: 06/04/2002
• Est. Priority Date: 09/05/1997
• Status: Expired due to Term

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Nesheim, William A., Hagersten, Erik E., Guzovskiy, Aleksandr, Jackson, Christopher J.
• Primary Examiner(s): Kim, Matthew
• Assistant Examiner(s): TZENG, FRED
• Application Number:
  US09/148,734
• Time in Patent Office: 1,369 Days
• Field of Search: 711/141, 711/147, 711/148, 711/150, 711/152, 711/153, 711/156, 711/3, 711/165, 711/168, 711/205, 711/206, 711/216, 709/200, 709/237, 709/215, 712/10, 712/11, 712/20, 712/23, 712/234, 712/37, 710/129, 713/600, 714/4, 714/15, 714/48, 714/752
• US Class Current: 711/147
• CPC Class Codes: G06F 11/0712 in a virtual computing platform, e.g. logically partitioned systems
• CPC Class Codes: G06F 11/0724 in a multiprocessor or a multi-core unit multiprocessors per se G06F15/80
• CPC Class Codes: G06F 11/073 in a memory management context, e.g. virtual memory or cache management memory management G06F12/00; testing of static memory units G11C29/00
• CPC Class Codes: G06F 11/0766 Error or fault reporting or storing
• CPC Class Codes: G06F 12/0284 Multiple user address space allocation, e.g. using different base addresses interprocessor communication G06F15/163
• CPC Class Codes: G06F 12/0864 using pseudo-associative means, e.g. set-associative or hashing
• CPC Class Codes: G06F 12/1018 involving hashing techniques, e.g. inverted page tables
• CPC Class Codes: G06F 12/1072 Decentralised address translation, e.g. in distributed shared memory systems
• CPC Class Codes: G06F 2212/2542 Non-uniform memory access [NUMA] architecture
• CPC Class Codes: G06F 2212/272 Cache only memory architecture [COMA]

Abstract

In one embodiment, a multiprocessing computer system includes a plurality of nodes. The plurality of nodes may be interconnected through a global interconnect network which supports cluster communications. An initiating node may launch a request to a remote nodes memory. In the event of an error, an error status register of a system interface of the launching cluster node is set to indicate the occurrence of an error. The error may be the result of an access violation, or the result of a time-out occurrence in either the remote node or the initiating node. Various other errors may alternatively be reported. The system interface advantageously includes a plurality of error status registers, with a separate error status register provided for each processor included in the node. A process running on any of the processors of the node reads an error by issuing a transaction to a unique address, wherein the unique address is independent of the processor upon which the process is running. The unique address as well as the transaction ID indicative of the processor which is attempting to read an error status are used by the system interface to determine which of the plurality of error status registers to access.

Claims

1. A multiprocessing computer system comprising a plurality of processing nodes and a global bus network interconnecting said plurality of processing nodes, wherein a first node includes:

a plurality of processors;

a memory coupled to said plurality of processors through a local bus; and

a system interface for receiving transactions initiated by said plurality of processors on said local bus which are destined to remote nodes, wherein said system interface includes a plurality of error status registers configured to store information regarding errors associated with transactions conveyed upon said global bus network, wherein a separate error status register is provided for each of said plurality of processors, and wherein said system interface includes a request agent coupled to said plurality of error status registers.

2. The multiprocessing computer system of claim 1, wherein said request agent maintains information identifying an initiating CPU for global transactions that are pending.

3. The multiprocessing computer system of claim 2, wherein said request agent utilizes said information identifying said initiating CPU to select a particular error status register corresponding to said initiating CPU for storing a value indicative of an error within said particular error status register.

4. The multiprocessing computer system of claim 3, wherein said request agent is configured to write said value to said particular error status register in response to receiving an error message from a remote node indicating an error in a global transaction which corresponds to a local transaction initiated by said initiating CPU.

5. The multiprocessing computer system of claim 4, wherein a process running on any of said plurality of processors reads a designated error status register corresponding to said process by issuing a transaction to a unique address, wherein said unique address is independent of the processor upon which said process is running.

6. The multiprocessing computer system of claim 5, wherein said request agent is configured to utilize said unique address and a transaction ID received from said local bus indicating a particular processor which is attempting to read an error status to thereby determine which of said plurality of error status registers to access.

7. The multiprocessing computer system of claim 3, wherein said value is indicative of a time-out error conveyed by a remote node.

8. The multiprocessing computer system of claim 3, wherein said value is encodable to indicate a time-out error within said first node.

9. The multiprocessing computer system of claim 3, wherein said value is encodable to indicate an access violation.

10. The multiprocessing computer system of claim 1, wherein said system interface is configurable to operate in either a cluster mode or an SSM mode.

11. The multiprocessing computer system of claim 10, wherein said cluster mode or said SSM mode is set in accordance with a control register of said system interface.

12. A method for communicating errors in a multiprocessing system operating in a cluster mode including a plurality of processing nodes, said method comprising:

a first cluster node sending a request to a second cluster node;

said second cluster node replying to said first cluster node with an error; and

said first cluster node storing an indication of said error within a particular error status register, wherein said first cluster node includes a plurality of error status registers wherein a separate error status register is provided for each processor within said first cluster node, and wherein said system interface includes a request agent coupled to said plurality of error status registers, and wherein said particular error status register is dependent upon a particular processor within said first cluster node which initiated said request.

13. The method of claim 12 further comprising encoding a value to indicate said error within said particular error status register.

14. The method of claim 13 wherein said value is encodable to indicate a time-out error.

15. The method of claim 13 wherein said value is encodable to indicate an access violation error.

16. The method of claim 12 further comprising:

a process executing on a first processor of said first cluster node initiating a read operation to a unique error status register address to determine the existence of an error; and

using said unique address in addition to a transaction ID which identifies said first processor as issuing said read operation to determine said particular error status register to read.

17. A multiprocessing system including a plurality of processing nodes which operate in a cluster mode, said multiprocessing system comprising:

a first cluster node including means for sending a request to a second cluster node;

said second cluster node including means for replying to said first cluster node with an error; and

means for storing an indication of said error within a particular error status register;

wherein said first cluster node further includes a plurality of error status registers wherein a separate error status register is provided for each processor within said first cluster node, and wherein said first cluster node includes a request agent coupled to said plurality of error status registers, and wherein said particular error status register is dependent upon a particular processor within said first cluster node which initiated said request.

18. The multiprocessing system of claim 17 further comprising means for encoding a value to indicate said error within said particular error status register.

19. The multiprocessing system of claim 18 wherein said value is encodable to indicate a time-out error.

20. The multiprocessing system of claim 18 wherein said value is encodable to indicate an access violation error.

21. The multiprocessing system of claim 17 further comprising:

a process executing on a first processor of said first cluster node, wherein said process initiates a read operation to a unique error status register address to determine the existence of an error; and

means for selecting a particular error status register to read depending upon said unique address in addition to a transaction ID which identifies said first processor as initiating said read operation.